test_ner3

This model is a fine-tuned version of distilbert-base-uncased on the pv_dataset dataset. It achieves the following results on the evaluation set:

• Loss: 0.2983
• Precision: 0.6698
• Recall: 0.6499
• F1: 0.6597
• Accuracy: 0.9607

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.1106	1.0	1813	0.1128	0.6050	0.5949	0.5999	0.9565
0.0705	2.0	3626	0.1190	0.6279	0.6122	0.6200	0.9585
0.0433	3.0	5439	0.1458	0.6342	0.5983	0.6157	0.9574
0.0301	4.0	7252	0.1453	0.6305	0.6818	0.6552	0.9594
0.0196	5.0	9065	0.1672	0.6358	0.6871	0.6605	0.9594
0.0133	6.0	10878	0.1931	0.6427	0.6138	0.6279	0.9587
0.0104	7.0	12691	0.1948	0.6657	0.6511	0.6583	0.9607
0.0081	8.0	14504	0.2243	0.6341	0.6574	0.6455	0.9586
0.0054	9.0	16317	0.2432	0.6547	0.6318	0.6431	0.9588
0.0041	10.0	18130	0.2422	0.6717	0.6397	0.6553	0.9605
0.0041	11.0	19943	0.2415	0.6571	0.6420	0.6495	0.9601
0.0027	12.0	21756	0.2567	0.6560	0.6590	0.6575	0.9601
0.0023	13.0	23569	0.2609	0.6640	0.6495	0.6566	0.9606
0.002	14.0	25382	0.2710	0.6542	0.6670	0.6606	0.9598
0.0012	15.0	27195	0.2766	0.6692	0.6539	0.6615	0.9610
0.001	16.0	29008	0.2938	0.6692	0.6415	0.6551	0.9603
0.0007	17.0	30821	0.2969	0.6654	0.6490	0.6571	0.9604
0.0007	18.0	32634	0.3035	0.6628	0.6456	0.6541	0.9601
0.0007	19.0	34447	0.2947	0.6730	0.6489	0.6607	0.9609
0.0004	20.0	36260	0.2983	0.6698	0.6499	0.6597	0.9607

Framework versions

• Transformers 4.21.0
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1